Revert "remove max_tokens from the official version of gpt4-turbo"

[modood]

This reverts commit dd4648ed9a803568b839e2510ca01cf7f1c6f740.

The azure - gpt-4-turbo-2024-04-09 model still requires the max_tokens parameter, so it doesn't work anymore since I updated to v2.12.3.

I think it is imprecise to simply determine whether the preview string is included, and I hope to revert this commit until there is a more precise way to determine it.

[vercel[bot]]

@modood is attempting to deploy a commit to the NextChat Team on Vercel.

A member of the Team first needs to authorize it.

[github-actions[bot]]

Your build has completed!

[Preview deployment]()

[Dean-YZG]

This reverts commit dd4648e.

The azure - gpt-4-turbo-2024-04-09 model still requires the max_tokens parameter, so it doesn't work anymore since I updated to v2.12.3.

I think it is imprecise to simply determine whether the preview string is included, and I hope to revert this commit until there is a more precise way to determine it.

As you already know, the latest big version of nextchat will refactor users' model entities and provider entities, and at that time, nextchat will support the configuration of max_tokens parameters for any provider's model, currently, you can include this feature in your project, in the near future, the new big version of nextchat will be better for chat